Fig 1.
Scheme of patient sample utilization.
Two hundred and ten sputum samples were collected. Thirty-one samples were initially excluded, leaving 179 samples for flow cytometric analysis. An additional 15 samples were omitted from the final analysis because of either too few cells in general (which does not amount to a reliable profile) or too few macrophages (which casts doubt to the lung origin of the sample). Ultimately 132 high-risk and 32 cancer samples were fully analyzed.
Fig 2.
Viability dye labels squamous epithelial cells as dead cells.
Light scatter profiles of 5 μm, 20 μm, and 30 μm particles (A) and dissociated sputum cells (B), both using the same voltage settings. The red boxes in A and B indicate a gate to exclude small debris (the bottom, left corner) as well as cell aggregates (last SSC and FSC channels). C) Sputum cells labeled with the FVS510 viability dye. D) Unstained sputum cells. The red (live cells) and blue boxes (dead cells) in C and D indicate populations of interest of sorting. E) Back-gating of dead cells to visualize the scatter profile of these cells. F, G) Wright-Giemsa-stained cells on a cytospin. The scale bars indicate 20 μm. F) Dissociated sputum cells prior to cell sorting. G) Cells collected from the dead cell gate comprised mostly of squamous epithelial cells. H) Cells collected from the live cell gate showing a heterogenous mixture of leukocytes and non-leukocytes.
Fig 3.
Sputum-derived leukocytes include distinct sub-populations of macrophages.
A-E) Presented are cells selected through a size exclusion gate and a live cell gate as shown in Fig 1, as well as a doublet discrimination gate (not shown). A) Representative light scatter profile of unstained single, live sputum cells defining both the CD45+ (red) and CD45- gates (blue) for sorting and further analysis. B) Live, single sputum cells stained with the blood panel of antibodies. The red box defines the CD45+ sorting population. C) Wright-Giemsa-stained cells obtained by sorting the CD45+ population, showing various types of leukocytes (*). D) Sputum-derived leukocyte profile of FVS510-CD45+ cells from a different sample stained with the antibodies indicated on the x- and y-axes. The black boxes indicate the gates used to identify lymphocytes/granulocytes (1), as well as alveolar (2) and interstitial macrophages (3). E) Fluorescence minus one (FMO) controls with the same boxes identifying leukocyte subpopulations as defined in D. All FMO controls include the viability dye, CD45, and TCPP. Ei) FVS510-CD45+ cells stained with the isotype controls for the antibodies indicated on each axis. Eii) Unstained cells. Eiii) Sputum cells stained with the leukocyte antibody panel minus the CD66b, CD3, and CD19 antibodies. Eiv) Sputum cells stained with the leukocyte antibody panel minus the CD206 antibody. F, G) Wright-Giemsa-stained cytospins from the sorted CD45+ gate 2 (F) and gate 3 (G) populations. Scale bars indicate 20 μm in F and 10 μm in G. Cell types were confirmed by a pathologist. H) Cell size measurements of the sorted macrophage population shown in panel F (gate 2) and G (gate3). For each population at least 100 cells were measured. Presented is the average cell size (+ SD). **** p <0.0001.
Fig 4.
Sample adequacy and criteria for sample exclusion.
Shown are the total number of sputum cells (excluding SECs) of individual samples prior to antibody labeling. All adequate samples (n = 164) revealed > 0.05% macrophages (alveolar and interstitial combined). Not depicted are 18 adequate samples, whose sputum cell count exceeded 50 million cells. The inadequate samples (n = 15) either showed no alveolar macrophages or the combined events in the alveolar and interstitial macrophage gates were < 0.05%. A subset of the inadequate samples contained “too few cells” for a reliable profile (< 1000 CD45+ events), while the remainder included enough cells, though did not fulfill the QC macrophage criteria to consider them adequate samples. The red lines and numbers indicate the median cell count for each sample group.
Fig 5.
Sputum-derived non-leukocytes identifying epithelial cell populations.
All profiles present cells selected through a size exclusion gate and a live cell gate as shown in Fig 1, as well as a doublet discrimination gate (not shown). A) The same cell suspension used to sort CD45+ cells as presented in Fig 3, was used to sort the CD45- fraction (blue gate). B) Wright-Giemsa-stained CD45- cells. The arrow indicates a ciliated epithelial cell and arrowhead indicates a goblet cell. All images were confirmed by a pathologist. Scale bars indicate 20 μm. C) Live, single CD45- sputum cells, from a different sample, stained with the epithelial antibody panel. D) Fluorescence minus one (FMO) controls for the profile presented in C. FMO controls include viability dye, CD45, and TCPP. Di) Sputum-derived epithelial profile of FVS510-CD45- cells stained with the isotype controls for the antibodies used in C. Dii) Unstained sputum cells. Diii) FMO control FVS510-CD45- cells stained with EpCAM but without the panCK antibody. Div) FMO control FVS510-CD45- cells stained with panCK but without the EpCAM antibody.
Fig 6.
Sputum cells highly stained with TCPP represent unique subpopulations.
A, B) Defining sputum cell populations with different TCPP fluorescence intensities (FI). A). The dot plot displaying TCPP versus FITC/Alexa488 fluorescence (i.e., CD66b/CD3/CD19 in the leukocyte tube (Ai) and panCK in the non-leukocyte tube (Aii)) is used to define the TCPPHIGH cut-off. The TCPP versus PE-CF594 fluorescence plot (not shown) can also be used for this purpose but the cells with the highest FI for TCPP are easier to identify in the former. B) Representative histogram of the TCPP FI. The TCPPHIGH cut-off is taken from the gate shown in panel A. The TCPPLOW population is defined at the intersect when unstained sputum is overlaid with the TCPP-stained sample. The population with intermediate TCPP staining, TCPPIM, is defined as the population between the TCPPHIGH and the TCPPLOW populations. Each of the three TCPP populations identified in B are further analyzed in C–E. Row C represents the TCPPHIGH subset analysis, row D the TCPPIM analysis and row E the TCPPLOW analysis. The first profile of each row (column i) shows the light scatter profile of the respective TCPP subpopulation. The second profile of each row (column ii) shows the distribution of CD45 staining of the cells in the respective TCPP subpopulation. The CD45+ fraction (“+”) of each TCPP subpopulation is further analyzed in column iii, which shows the distribution of CD66b/CD3/CD19 staining versus CD206. The CD45- ("-") fraction of each TCPP subpopulation is represented in column iv, which shows the panCytokeratin versus EpCAM staining.
Fig 7.
Differences in sputum cell characteristics between cancer and high-risk sputum samples.
Depicted are the significant differences between cancer (CA; n = 32) and high-risk samples (HR; n = 132) resulting from the analysis described in Fig 5. Each dot (CA) and square (HR) represent one sample. A) the TCPPHIGH population in cancer samples displays a smaller SSC than the TCPPHIGH population in high-risk samples (** p <0.01). B) In cancer samples, the proportion of EpCAM+panCK+ cells in the CD45- fraction of the TCPPHIGH subpopulation is larger than in the corresponding CD45- fraction in high-risk samples (** p < 0.01). C) The mean fluorescence intensity (MFI) of EpCAM in TCPPHIGH CD45-EpCAM+panCK+ cells is higher in cancer samples than in the corresponding cellular subset of high-risk samples (* p < 0.05). The red lines indicate the median values for each sample group.